lumbar load as relevant spinal injury metrics comparing ...comparing the use of dynamic response...

Registration No.

-Technical Report-

U.S. Army Tank Automotive Research,Development, and Engineering CenterDetroit ArsenalWarren, Michigan 48397-5000

Ravi Thyagarajan1, Jaisankar Ramalingam1, Kumar B Kulkarni2

1 US Army TARDEC, Warren, MI2 ESI-US Inc, Troy, MI

UNCLASSIFIED: Distribution Statement A

Approved for Public Release

Occupant-Centric Platform (OCP) Technology-Enabled

Capabilities Demonstration (TECD)

Comparing the Use of Dynamic Response Index (DRI) and

Lumbar Load as Relevant Spinal Injury Metrics

Presented at the ARL Workshop on Numerical Analysis of Human and Surrogate Response to Accelerative Loading, Jan 09 2014

09 January 2014

UNCLASSIFIED

UNCLASSIFIED

Comparing the Use of Dynamic Response Index (DRI) and Lumbar Load as Relevant Spinal Injury Metrics

UNCLASSIFIED: Distribution Statement A. Approved for Public Release 2 | P a g e

REPORT DOCUMENTATION PAGE Form Approved

OMB No. 0704-0188 Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing this collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS.

1. REPORT DATE (DD-MM-YYYY)

09 JANUARY 2014 2. REPORT TYPE

Brief at ARL Workshop on Accelerative Loading 3. DATES COVERED (From - To)

10/01/2013 – 01/10/2014

4. TITLE AND SUBTITLE

5a. CONTRACT NUMBER

W56HZV-08-C-0236

W56HZV-08-C-

0236


5b. GRANT NUMBER

5c. PROGRAM ELEMENT NUMBER

6. AUTHOR(S)

5d. PROJECT NUMBER

Ravi Thyagarajan, Jaisankar Ramalingam, Kumar B Kulkarni 5e. TASK NUMBER

WD0046 Rev 4

5f. WORK UNIT NUMBER

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)

AND ADDRESS(ES)

8. PERFORMING ORGANIZATION REPORT NUMBER

ESI-US Inc

888 W Big Beaver Road #402

Troy MI 48084

TARDEC/Analytics

6501 E 11 Mile Road

Warren MI 48397

9. SPONSORING / MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)

Sponsors: Monitor: BP3I, HPCMO, ISABEL, TARDEC, OCP, TECD Blast Institute and HPC Mod Office, TARDEC/Analytics Aberdeen, MD

6501 E 11 Mile Road

11. SPONSOR/MONITOR’S REPORT

OCP TECD Program, Warren MI 48397

NUMBER(S) #23785 (TARDEC) Warren, MI #24373 (TARDEC)

12. DISTRIBUTION / AVAILABILITY STATEMENT

UNCLASSIFIED: Distribution Statement A. Approved for Public Release, Unlimited Distribution

13. SUPPLEMENTARY NOTES

14. ABSTRACT

The two most commonly used injury criteria for Spinal injuries today are Dynamic Response Index (DRI) related to structural accelerations, usually of the seat pan, or even more directly, lumbar force measurements taken within the Hybrid-III ATD as the evaluation criterion. With respect to continued use of these two criteria for spinal injuries, this report examines the following aspects in detail: 1) Any existing correlation between Peak Lumbar loads and DRI for un-encumbered occupants, in the whole blast loading regime or at least within different loading regimes 2) Re-evaluate (1) for encumbered occupants, that is, with heavier upper torsos 3) Potential changes to DRI calculations and Injury Assessment Reference Value (IARV) thresholds for encumbered occupants 4) General discussion on continued use of DRI as a design criterion for spinal injuries given the availability of the more direct Lumbar load from fully encumbered ATDs in underbody blast testing. 15. SUBJECT TERMS

DRI, Lumbar Load, Blast, LSDYNA, MADYMO, occupant, injury, pelvic, IARV, Encumbered, Mertz, Geertz, Brinkley, lumbar, spine, Anton 16. SECURITY CLASSIFICATION OF:

17. LIMITATION OF ABSTRACT

18. NUMBER OF PAGES

19a. NAME OF RESPONSIBLE PERSON Ravi Thyagarajan

a. REPORT

Unlimited b. ABSTRACT Unlimited

c. THIS PAGE Unlimited

Unlimited 23

19b. TELEPHONE NUMBER (include area

code) 586-282-6471

Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std. Z39.18



TANK-AUTOMOTIVE RESEARCH

DEVELOPMENT ENGINEERING CENTER Warren, MI 48397-5000

Occupant-Centric Platform (OCP) Technology-Enabled Capabilities Demonstration

(TECD)

09 January 2014

Comparing the Use of Dynamic Response

Index (DRI) and Lumbar Load as Relevant

Spinal Injury Metrics

By

Ravi Thyagarajan1, Jaisankar Ramalingam

1, Kumar B Kulkarni

2

1 US Army TARDEC, Warren, MI 2 ESI-US Inc, Troy, MI

This is a reprint of the brief presented under the same title during the ARL

Workshop on “Numerical Analysis of Human and Surrogate Response to

Accelerative Loading”, Jan 7-9, 2014 in Aberdeen, MD.



Distribution List

Mr. Sudhakar Arepally, Associate Director, Analytics, US Army TARDEC

Dr. Pat Baker, Director, ARL/WMRD, Aberdeen, MD

Mr. Craig Barker, Program Manager, UBM/T&E, SLAD, US Army Research Lab

Mr. Ross Boelke, OCP-TECD PM, TARDEC/GSS

Mr. Robert Bowen, ARL/SLAD, Aberdeen, MD

Dr. Kent Danielson, Engineer Research and Development Center (ERDC), Army Core of Engineers

Mr. Paul Decker, Deputy Chief Scientist, US Army TARDEC

Mr. Matt Donohue, DASA/R&T, ASA-ALT

Ms. Nora Eldredge, WMRD, US Army Research Lab

Mr. Ed Fioravante, WMRD, US Army Research Lab

Mr. Ami Frydman, WMRD, US Army Research Lab

Mr. Mark Germundson, Deputy Associate Director, TARDEC/GSS

Mr. Neil Gniazdowski, WMRD, US Army Research Lab

Dr. David Gorsich, Chief Scientist, US Army TARDEC

Dr. Chris Hoppel, WMRD, US Army Research Lab

Mr. Jeff Jaster, Deputy Associate Director, TARDEC/GSS

Mr. Steve Knott, Deputy Executive Director, TARDEC/GSEAA, US Army TARDEC

Mr. Jeff Koshko, Associate Director, TARDEC/GSS, UA Army TARDEC

Mr. Joe Kott, OCP-TECD Deputy PM, TARDEC/GSS

Mr. Dick Koffinke, Survivability Directorate, US Army Evaluation Center

Dr. Scott Kukuck, PM/Blast Institute, WMRD, US Army Research Lab

Dr. David Lamb, STE/Analytics, US Army TARDEC

Mr. Mark Mahaffey, ARL/SLAD, Aberdeen, MD

Dr. Tom McGrath, US Navy NSWC-IHD

Mr. Tony McKheen, Associate Director/Chief Integration Engineers, US Army TARDEC

Mr. Kirk Miller, OCP-TECD Standards and Specifications, TARDEC/GSS

Mr. Micheal O’Neil, MARCOR SYSCOM, USMC

Mr. Mark Simon, Survivability Directorate, US Army Evaluation Center

Dr. Paul Tanenbaum, Director, ARL/SLAD, Aberdeen, MD

Dr. Doug Templeton, TARDEC/GSS

Mr. Pat Thompson, US Army Testing and Evaluation Command (ATEC)

Mr. Madan Vunnam, Team Leader, Analytics/EECS, US Army TARDEC

TARDEC TIC (Technical Information Center) archives, US Army TARDEC

Defense Technical Information Center (DTIC) Online, http://dtic.mil/dtic/

http://dtic.mil/dtic/


Lumbar Load as Relevant Spinal Injury Metrics UNCLASSIFIED: Distribution Statement A. Approved for Public Release 1

Comparing the Use of Dynamic

Response Index (DRI) and Lumbar

Load as Relevant Spinal Injury

Metrics

Ravi Thyagarajan

Jai Ramalingam

Kumar Kulkarni TARDEC/Analytics

Workshop on Numerical Analysis

of Human and Surrogate Response

to Accelerative Loading Army Research Laboratory (ARL)

Aberdeen, MD

Jan 7-9, 2014

UNCLASSIFIED: Distribution Statement A. Approved for Public Release


Lumbar Load as Relevant Spinal Injury Metrics UNCLASSIFIED: Distribution Statement A. Approved for Public Release

OUTLINE

• Mechanical and Injury Models for DRI

• Mechanical and Injury Models for Lumbar Load

• DRI and LL: Temporal Behavior

• M&S Model Descriptions

• Behavior of Peak Compressive LL vs. DRI Cross-plots

• Proposal for Mechanical Model for Encumbered DRI

• Known Issues with DRI

• Summary / Conclusions

2



Dynamic Response Index (DRI) –

Mechanical Model

• Simple lumped mass parameter model

(single spring-mass-damper) to simulate the

biomechanical response of the human upper

body/vertebral column/pelvis [2,3]

• Values of m, k, c (and thus wn, z) were

derived by compressive strengths of

individual vertebrae [1], and load-deflection

curves [4]

• Values established for a representative

population of Air Force pilots with a mean

age of 27.9 years [3]

m = 34.51 kg

k = 9.66E04 N/m

c = 818.1 Nsec/m

• wn = 52.9 rad/s, and z = 0.224

• Lumbar Force = k*d

• Maximum Lumbar Force = k*dmax

3

(d=D1-D2)

wn = sqrt(k/m)

z = c/(2*sqrt(m*k))

Input accel

Input disp

Disp of mass

Input disp - Mass disp

Maximum Lumbar Force, when normalized by the weight m*g, is called DRI

• Normalized Lumbar Force

= (k*dmax) / (mg) = w2n *dmax/g

UPPER BODY

PELVIS

LUMBAR

D2

D1

d dmax



Dynamic Response Index (DRI) –

Injury Risk Model

• During World War II, Geertz generated data

on compressive vertebral strengths either with

individual vertebrae or vertebral complexes of

PMHS between 19 and 46 years old [1]

• Stech and Payne [3] used the above to relate

the DRIz to an injury risk of 50% vs age, and

for an average age of 27.9 of Air Force pilots,

estimated a DRIz of 21.3 (7220N/1622 lbf).

Brinkley used a normal distribution around

this to set up the laboratory data curve [5]

• DRI value of 17.7 leads to a 10% risk of

spinal injury (corresponds to 5992 N/1346 lbf)

• Injury model based on the laboratory data

curve has the pelvis as point of initiation, so

as far as possible, the pelvic acceleration

rather than the seat acceleration should be

used to calculate the DRI [13]

4

• Quasi-static testing on PMHS specimens led to a 10% risk of spinal injury for DRI = 17.7

• Underlying principles for DRI model are based on lumbar load

Spinal Injury Risk Calculated from Laboratory and Operational Data valid for AIS 2+ Injuries [3,5] .

10%

17.7



Compressive Lumbar Load (LL) –

Mechanical Model

• Development of anthropomorphic test devices (ATD)

and subsequent addition of load transducers in them

represents a revolutionary increase in capability [20]

• Curved lumbar spine is incorporated to replicate

typical seated automotive occupant positions, also

used in military vehicle applications

• Three-Axis Lumbar Spine Load Cell measures time-

dependent forces/moment at desired sampling rates

during blast/crash

5

Lumbar Force measured here is a “direct” representation of lumbar response/injury

HYBRID-III ATD [23]

Part 572 Hybrid-III Lower Torso Assembly [23]

PELVIS

LUMBAR SPINE

• Lumbar load cell did not adversely affect

measured accelerations and forces, nor

modify the spinal flexural characteristics [17]

• ATDs capable of producing reproducible

results in greater detail under controlled

testing conditions

• Biofidelic enhancements to the Hybrid III

design were made which support its use in

predicting human injury during high-speed

dynamic events [20]


Lumbar Load as Relevant Spinal Injury Metrics UNCLASSIFIED: Distribution Statement A. Approved for Public Release 6

Different approaches, but they lead to similar injury criteria for Compressive Lumbar Load

Compressive Lumbar Load (LL) –

Injury Risk Model

Source Chandler [7,10] Tremblay [24], Ripple [25] Mertz [18,21,22,12]

Approach • Proposed for aircraft seats

• Derived a compression force criterion

by correlating the DRI and maximum

compression force measured on a H-II

lumbar load spine cell in 12 tests

• 1500 lb / 6675 peak value corresponds

to a DRI of ~19

• Adopted by FAA Regulations in Title

49/CFR 572

• Proposed by Tremblay based on

Ripple and Mundie’s paper, but that

paper doesn’t specify any tolerance

values, so not clear on origin

• NATO RTO-TR-HFM-090 suggests that

these criteria arise from Mertz criteria

[19] for a scaling factor of 3.4-3.8,

which also does not match the paper.

• Perhaps Tremblay meant to refer to

Alem [6], who also refers to a factor 3.4

that was used on Mertz’s neck data in

estimating 6675 N for peak lumbar load

• Tolerance curves for compressive

neck loading in high school football

players and the adult populace using

a H-III ATD outfitted with a football

helmet impacted by a tackling block

• Scaling factor from neck to lumbar

based on waist and neck dimensions

• Limiting force rationale (ratio applied

to large-duration value); more

conservative in mitigating lumbar

spine injuries

ATD H-II Straight Lumbar Spine Unknown H-III Curved Lumbar Spine

Peak LL

Criterion (C)

1500 lb / 6675 N 1500 lb / 6673 N 1258 lb / 5598 N

r2~ 0.4

30 ms

6673 N

3800 N

0

2000

4000

6000

8000

0 10 20 30 40 50

0

2000

4000

6000

8000

0 10 20 30 40 50

1100

5598 5598

Neck Lumbar Spine

IARV for Compression Force

Duration of loading over given force level (ms)

Scaled by

5.089

Duration of loading over given force level (ms)

1 2

3

4

5

6

7

8

IARV for Compression Force

10 20 40 50



DRI and LL: Temporal Behavior

• Same pulse applied to two slightly different (both rigid) seat configurations in M&S

• Direct value of LL for C1 is 10.5 KN, indirect value (from DRI) is 8.5 KN; In addition, time of

occurrence of the peak is quite different (8.7 ms after table peak for LL vs. 35.9 ms for DRI)

• The larger shift means that later changes in Pelvic accel will still affect DRI, but not peak LL

• Peak LL for C2 shifted by 2.8 ms vs. C1, and value higher by 13%

• Pelvic acceleration peak for C2 shifted by 2.8 ms, only 1% less, but wider (based on 7 ms clip)

• DRI for C2, calculated from pelvis acceleration, is shifted by 1.8 ms, value higher by only 8%

7

DRI does not track direct LL well in the time domain and can be affected by minor late data



M&S Model Description - MADYMO

• A triangular blast wave pulse was applied to the vertical drop tower/sled.

• For unencumbered occupant: A total of eleven duration levels are studied; from 2 ms to 60 ms.

At each DT, peak deceleration was varied from 10g to 1200g in 10g increments up to the point

when ∆v reached ~15m/s, where ∆v =0.5*Peak acceleration*DT (230 runs for each seat type)

• ~30 kg of encumbered PPE mass on the typical AM50 soldier was lumped on the upper torso,

and 31 simulations were run covering Peak acceleration between 10-360g, DT between 5-15

ms (corresponding to a ∆v between 1.5-12 m/s) for encumbered occupants

8

Systematic study was performed on MADYMO ATD setup for a large sample of input pulses

MADYMO Dynamic simulation model including

Q-version of AM50 H-III ATD

In addition to Rigid Seat, two other EA seats (4 and 8 KN

limiting force) were also included

230 Triangular blast inputs were used for each of the 3 seats



M&S Model Description - LSDYNA

• A triangular blast wave pulse was applied to the vertical drop tower/sled.

• Only rigid seats were used in the LS-DYNA simulations, and the Humanetics version [23] of the

LS-DYNA ATD (military version) were used

• For encumbered occupant studies, the vest and helmet were modeled in FEA using finite

elements. The remaining PPE mass on the upper body of a typical AM50 encumbered soldier

(~30kg - mass of vest) was lumped on the vest

• 31 simulations were run covering Peak acceleration between 10-360g, DT between 5-15 ms

(corresponding to a ∆v between 1.5-12 m/s) for both unencumbered and encumbered occupants

9

A reduced set of simulations were performed on LS-DYNA ATD setup

Encumbered AM50 H-III Occupant

Helmet

1.4 kg

Upr Body

PPE

29.4 kg

AM50 ATD

78.1 kg

Boots

2.3 kg

Un-Encumbered AM50 H-III Occupant

Triangular pulse applied to table of magnitude Ap and time duration DT



Behavior of Peak Compressive LL vs. DRI

• DRI and LL test data have been obtained for restrained occupants (usually encumbered) from

a multitude of vehicles of different sizes and weights subjected to underbody mines of different

sizes, positioned in different seats in different vehicle positions and configurations.

• While amount of scatter is reduced for M&S data, it is clear that there is a lack of a general

overall governing relationship between DRI and Peak LL.

• When some other factors are also included, for example, only data for a seat type and a

reduced range of DT, some patterns can be discerned in the M&S data.

• One interesting observation is that based on previously described IARVs, for 94% of the

samples in test and 89% in M&S, DRI and LL both predict the same outcome (incapacitation,

or no incapacitation)

10

Data from underbody mine tests (~1200 samples) Data from MADYMO M&S (~700 samples)



Unencum Encum #1 Encum #2

m m m+M m+M

k k k k

c c c c

w2n k/m k/m k/(m+M)

DRI

= (k*dmax)

mg

= (k*d’

max)

mg

= (k*d’

max)

(m+ M)g

IARV 17.7 17.7 17.7

(1 + M/m)

RI =

DRI /

IARV

k*dmax

17.7*mg

k*d’max

17.7*mg

k*d’max

17.7*mg

Mechanical Model for DRI of Encumbered

Occupant (DRI’)

11

• The Mass quantity in the DRI SDOF calculator MUST be increased to compensate for the

encumbrance (actual factor to be used (<1) on added mass is under review)

• Approach #1 is strongly preferred since the familiar IARV values (17.7) are still the same

M corresponds to the weight to be added to m in DRI calculator to account for encumbrance on upper body of occupant. It is usually only a fraction of the actual physical weight of the encumbrance.

Unencum Encum #1 Encum #2

m, kg 34.51 64.91 64.91

k, N/m 9.66E4 9.66E4 9.66E4

c, Ns/m 818.1 818.1 818.1

w2n 2799.2 2799.2 1488.2

DRI 285.34*

dmax

285.34*

d’max

151.86*

d’max

IARV 17.7 17.7 9.4

RI =

DRI /

IARV

16.1*

dmax

16.1*

d’max

16.1*

d’max

Example: M=30.4 kg

’ ’ ’



DRI for Encumbered Occupants (DRI’) -

Example

12

Depending on how the DRI calculator is coded and which normalization factor is used, the

DRI value will be different and must be compared against the right IARV

Let’s calculate DRI for unencumbered (m=34.51

kg) and encumbered (m=64.91 kg) occupants

based on the SDOF system, i.e, assuming that

an added mass of 30.4 kg to the upper body

fully affects lumbar load and DRI.

The analysis is being done for triangular pelvic

accelerations all of fixed duration 5 ms, but the

amplitude is allowed to vary from 25-300g

5992 N

17.7

9.4

131g 188g

131g 188g 241g

• From the max allowable lumbar load of 5992 N,

it can be seen that a max pelvic acceleration of

188g and 131g can be withstood, for the

unencumbered and encumbered occupants.

• Depending on the normalization constant used

in determination of the encumbered DRI, the

correct corresponding IARV must be used to get

consistent and accurate results.

• Because the lumbar load (k*d) is uniquely

defined, it is a good idea to verify that DRI

results are consistent with lumbar load.

Pelvic acceleration, g

Pelvic acceleration, g



• For a specific seat and limited DT range, DRI-LL relationship is linear in both tests and M&S

• For the LS-DYNA results, when the pelvic accels from increased upper body weights are run

through the DRI calculator without changing the mass, the DRIs are seen to drop (red curve),

which is not realistic, since the lumbar loads increased. Using DRI calculations as per #1, shifts

to green curve, indicating higher DRIs as expected.

• For MADYMO results, because all the added weight affects pelvic accel, the DRI curve moves

from blue to red, and increases even further as per #1. This indicates that a smaller factor (<1)

needs to be applied to the physical added mass in order to accurately capture vest separation.

13

DRI (properly calculated) and LL both go up for encumbered occupants

LL vs DRI behavior from drop tower tests (left), LSDYNA M&S (middle) and MADYMO (right) M&S

Behavior of Peak Compressive LL vs. DRI

Peak Acceleration,g Duration, ms ∆V, m/s

M&S 10 - 360 5 - 15 1.5 - 12

Test 3 - 285 5, 20 3 -7



DRI - Applicability to Ejection Seats

• Stech and Payne also presented the injury

risk for operationally experienced non-fatal

spinal injuries in ejection seat tests, shown

as operational data curve [3, 5?]

• F-4 operational data [5] does not match the

injury trend. For the F-4 DRI value of ~19,

the operational data curve yields 9% risk of

injury, not the observed 34% in reality.

• DRI-Injury Rate Relationship is only valid for

misalignments of the seat with respect to the

catapult direction < 5 degrees, which was not

true for the F-4 seat

• In a survey of 223 ejections by British aircraft

pilots over 1968-83, Anton [8] found a poor

agreement between the incidence of spinal

fracture and the DRI for ejections from 5 out

of 6 ejector seats and concluded that

predictors such as DRI “have no apparent

practical utility” [11]

14

Ejector Seat Data raises doubts as to suitability of the DRI as an injury measure [11]

9%

~19

34%

Spinal Injury Risk Calculated from Laboratory and Operational Data valid for AIS 2+ Injuries [3,5] .



DRI – Other Known Issues

• The curved lumbar spine is incorporated in H-III ATD to replicate typical seated automotive occupant

positions in military vehicle applications. This results in a misalignment of the accelerometer axes and

the lumbar spine by about 210. The DRI, which by definition, assumes a straight lumbar spine,

deviates in this case from its intent to be an indicator of lumbar force.

• The DRI represents a whole body motion criterion which represents a load criterion instead of an

injury criterion. Load criteria are based on physical parameters which specify an external load on the

human body (e.g. footplate intrusion), whereas injury criteria are established with physical parameters

which describe the biomechanical response of the human body or its surrogate [14]. Neck and

Lumbar loads are examples of injury criteria.

• The DRI model is based on unconstrained motion of a single constant reaction lumped mass.

Restraint systems impede the vertical motion, especially for mine blast seats which extend the loading

duration. The reaction mass is increasingly constrained during the duration of blast response.

Because the model treats the whole body as a lumped mass, the seat geometry and restraints used in

the test data are critical to achieve the same results [9].

• As noted in [13], the physical parameter which affects fracture is always force. Using a model which is

based on another physical parameter causes less accuracy and can lead to contradictory results.

• Even though the DRI model is based on single-degree-of-freedom vibration, it has been found [11]

that even for continuous vibration, at frequencies > 8.4 Hz, the response tends to decrease in

proportion to freq2, so the predicted stress on the spine decreases at 12 dB per octave. Consequently,

when the DRI model is used for continuous sinusoidal motion, it erroneously indicates that excessively

high accelerations are permissible at high frequencies.

• The DRI model lacks fidelity in regards to gender, weight, anthropometrics and age.

15



DRI – Other Known Issues (contd)

• In several mine protection trials, seat acceleration data have shown to have a high variation and a

lack of reproducibility [13]. Although the DRI SDOF system is comparable to a filter and tends to

smooth out the input acceleration, the variation of the seat acceleration input has a negative impact on

the reproducibility of the DRI data. An apparent advantage that the DRI measure has, in that it can still

be computed when ONLY the seat acceleration data is available, tends to get neutralized by the

above finding.

• Additional helmeted and vest masses may cause the natural frequency and damping characteristics of

the human to change, invalidating the model [9]

• The assumption of linearity of the DRI model is highly unrealistic. It has been shown [16] that the

frequency characteristics of the upper human body are distinctly different at low and high amplitude

accelerations. Furthermore, the same paper also points out that in vitro compression testing of L1-L2

spinal units have indicated a non-linear force-displacement curve. Such non-linear characteristics

have been and can be easily incorporated into the ATD models and hardware for determination of

more accurate lumbar loads.

• The simplified assumption of a single mass, stiffness, and damping value, and reliance on pelvic or

seat acceleration over the full time duration, leads to the undesired behavior of the DRI being affected

by late peaks and valleys in the input acceleration, significantly after the effect of the blast load has

already occurred.

• The DRI, by the nature of its very definition, has limited number of variables that can be changed to

account for any new research findings on lumbar spine behavior. In contrast, the continued

development of end-to-end, full system underbody blast tools [15] and the determination of the LL

from an detailed ATD provides a much better “upgrade path” to accommodate new emerging data and

predict lumbar spine injuries.

16



Summary / Conclusions

• DRI has the attraction of being an apparently simple, tangible model which clearly had high utility

before the advent of detailed ATDs that could produce reproducible results in controlled testing.

• While DRI can be calculated when only seat accelerations are available and indeed may be the only

injury measure that can be calculated in such a case (like when an ATD is not used), the consistency

and usefulness of such data is highly questionable due to the variability in the seat accelerations.

• Human responses are highly nonlinear, and to expect a simple linear model such as DRI to be capable

of responding accurately to a wide range of shock amplitudes is highly unrealistic.

• DRI and LL responses are both dynamic, and the peak values may even be in the ball-park, but DRI

lags far behind as to when the peak occurs due to the use of only one frequency characteristic. This

can lead to unrealistic consequences where later changes in pelvic acceleration can affect the DRI.

• There is a lack of any kind of overall general correlation between DRI and LL.

• Requiring pelvic accelerations for accurate DRI calculations means ATD is required. In which case,

the lumbar load can be directly measured and compared against its IARV.

• Calculating DRI for encumbered occupants can be tricky in that while it is clear that the increased

mass increases the lumbar load, what factor to use on the actual physical mass is still not clear. Also,

it is recommended that if DRI is used at all, that it be determined using the standard normalization

constant so that the familiar DRI values are still preserved.

• The availability of force-based IARV injury criteria on direct measurements such as lumbar load,

makes them highly attractive as candidates for incapacitation assessment for the lumbar region.

17

Too simple, Too many assumptions, Too many questions…. DRI had an important role

50 years ago in the evolutionary timeline, but has the since largely outlived its utility

….. Time to move to a more “direct” injury measure (Lumbar Load from detailed ATDs)



GLOSSARY / ACRONYMS

18

AIS Abbreviated Injury Scale AM50 American Male 50th Percentile APG Aberdeen Proving Grounds, Maryland ARL Army Research Laboratory ATD Anthropomorphic Test Device ATEC Army Test and Evaluation Center COTS Commercial-off-the-Shelf DOF Degree-of-Freedom FEA/FEM Finite Element Analysis/Method g acceleration due to gravity H-II / H-III Hybrid-II or Hybrid-III ATD kg kilogram, unit of mass; 1kg ~ 2.204 lb lb/lbf pounds, pounds of force; 1lbf ~ 4.45 N IARV Injury Assessment Reference Value LL Lumbar Load LSDYNA COTS structural dynamics software from LSTC, CA LSTC Livermore Software Technology Corporation, CA ms msec, milliseconds, unit of time (1 ms = 0.001 second) M&S Modeling & Simulation MADYMO MAthematical DYnamic MOdels (COTS software from TNO) N Newtons, unit of force, 1 N ~ 0.22472 lbf OCP Occupant-Centric Platform PMHS Post-mortem Human Specimens R&D Research & Development RDECOM Research, Development and Engineering Command RI Relative Injury Index = Injury Value / IARV SimBRS Simulation-Based Reliability and Safety SDOF Single Degree-of-Freedom SLAD Survivability and Lethality Analysis Directorate in ARL TARDEC Tank Automotive Research, Development and Engineering Center T&E Test & Evaluation UBM Underbody Blast Modeling/Methodology WMRD Weapons and Materials Research Directorate in ARL



ACKNOWLEDGMENTS / DISCLAIMER

19

DISCLAIMER

Reference herein to any specific commercial company, product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or the Dept. of the Army (DoA). The opinions of the authors expressed herein do not necessarily state or reflect those of the United States Government, the DoD, or U.S. Army TACOM Life Cycle Command and shall not be used for advertising or product endorsement purposes.

ACKNOWLEDGMENTS

The authors would like to thank the Blast Protection for Platforms and Personnel Institute (BP3I) program managed by ARL/WMRD and the HPC Modernization Office, for the partially funding of this project. Thanks also to the OCP TECD Program managed by TARDEC, Warren, MI for funding support. The authors also gratefully acknowledge physical test data provided by Mr Craig Barker/Mr Brian Benesch of ARL/SLAD, and Mr Ami Frydman of ARL/WMRD. We express our gratitude to Dr. Harold (Bud) Mertz for reviewing this material and providing valuable feedback.

This material is based on R&D work partially supported under Contract No. W56HZV-08-C-0236, through a subcontract with Mississippi State University (MSU), and was performed for the Simulation Based Reliability and Safety (SimBRS) research program. Any opinions, finding and conclusions or recommendations in this paper are those of the authors and do not necessarily reflect the views of the U.S. Army TACOM Life Cycle Command.



REFERENCES

20

1. Ruff, S., Brief Acceleration: Less than One Second, German Aviation

Medicine in World War II, Vol. I, Chapter VI-C, Dept of the Air Force, 1950

2. Latham, F., A Study in Body Ballistics, Seat Ejection, Proceedings of the

Royal Society of London, Series B – Biological Sciences, Vol. 147, pp. 121-

139, 1957

3. Stech, E.L. and Payne, P.R., Dynamic Models of the Human Body,

Aerospace Medical Research Laboratory, Wright Patterson Air Force Base,

Ohio, USA, 1969

4. Yorra, A.J., The Investigation of the Structural Behavior of the Intervertebral

Discs, Master Thesis, Massachusetts Institute of Technology, 1956

5. Brinkley, J.W. and Shaffer, J.T., Dynamic Simulation Techniques for the

Design of Escape Systems: Current Applications and Future Air Force

Requirements, Symposium on Biodynamic Models and their Applications,

Report No. AMRL-TR-71-29, Aerospace Medical Research Laboratory,

Wright-Patterson Air Force Base, Ohio, USA, 1970

6. Alem, N.M., Acceleration Injuries: Assessment Methods and Criteria,

Protection of Wheeled Vehicle Occupants from Landmine Effects, WTP 1,

KTA 1-29, Melbourne Australia, 1996



REFERENCES

21

7. Chandler, R.F., Human Injury Criteria Relative to Civil Aircraft Seat and

Restraint Systems, 851847, Society of Automotive Engineers, PA, USA, 1988

8. Anton, D.J., The incidence of spinal fracture on Royal Air Force ejections

1968-1983, Report No. 529, Aircrew Equipment Group, Royal Air Force,

Institute of Aviation Medicine, Farmborough, UK, 1986

9. Caldwell, E., Gernhardt, M., Somers, J.T., Younker, D., Newby, Evidence

Report: Risk of Injury due to Dynamic loads, Human Reseach Program,

Human Health and Countermeasures Element, NASA –Johnson,TX, 2012

10. Spink, R.J., Injury Criteria for the Analysis of Soldier Survivability in

Accelerative Events, ARL-TR-6121, 2012

11. Griffin, M.J., Handbook of Human Vibration, Academic Press, London, ISBN

0-123030412, 1990

12. Sanders, D., Hanlon, E., Howard, H., Mertz, H., Wodzinski, C., Enhanced

Injury Assessment Reference Value Report (DRAFT), OCP-TECD Report,

Feb 2013

13. Final Report of HFM-090 Task Group 25, Test Methodology for Protection of

Vehicle Occupants against Anti-Vehicular Landmine Effects, NATO RTO

TECHNICAL REPORT TR-HFM-090, April 2007



REFERENCES

22

14. Dosquet, F., Lammers, C., Soyka, D., Hennemann, M., Hybrid III Thoraco-

lumbar Spine Impact Response in Vehicular Protection Applications for

Vertical Impacts, PASS, 2010

15. Thyagarajan, R., End-to-end System level M&S tool for Underbody Blast

Events, 27th Army Science Conference, Army Technology Showcase,

Orlando, FL, Nov 29 – Dec 2. DTIC Report # ADA550921, 2000

16. Nicol, J.J., Modeling the Dynamic Response of the Human Spine to

Mechanical Shock and Vibration Using an Artificial Neural Network, Ph. D.

Thesis, Simon Fraser University, Aug 1996

17. Schoenbeck, A., Forster, E., Rapaport, M., Domzalski, L., Impact Response of

Hybrid III Lumbar Spine to +Gz Loads, 981215, Society of Automotive

Engineers, Warrendale, PA, USA, 1998

18. Mertz, H.J., Irwin, A. L., Prasad, P., Biomechanical and Scaling Bases for

Frontal and Side Impact Injury Assessment Reference Values, Stapp Car

Crash Journal, 2003-22-0009, Vol. 47, pp. 155-188, October 2003

19. Mertz, H.J., Jr. and Patrick, L.M., Investigation of the Kinematics and Kinetics

of Whiplash, 670919, Society of Automotive Engineers, Warrendale, PA, USA,

1967



REFERENCES

23

20. Mertz, H.J., Injury Assessment Values Used to Evaluate Hybrid III Response

Measurements, GM submission USG 2284 to NHTSA Docket 74-14/Notice 32,

1984. Later reprinted in Hybrid III: The First Human-Like Crash Test Dummy,

SAE PT-44, SAE, Warrendale, PA,1994

21. Mertz, H.J., Jarrett, K., Moss, S., Salloum, M. and Zhao, A., The Hybrid III 10-

Year-Old Dummy, Stapp Car Crash Journal, Vol. 45, Warrendale PA, 2001

22. Mertz, H.J., Hodgson, V.R., Thomas, L.M., Nyquist, G.W., An Assessment of

Compressive Neck Loads Under Injury-Producing Conditions, The Physicial

and Sports Medicine, pp. 95-106, November 1978

23. Humanetics Innovative Solutions, Hybrid-III 50th Male Dummy, 78051-218-H,

FMVSS208, 49CFR Part 572, Subpart E, Hybrid-III 50th Male Dummy Parts

Catalog, Rev. 4, July 2013

24. Tremblay, J., Bergeron, D.M., Gonzalez, R., Protection of Soft-Skinned

Vehicle Occupants from Landmine Effects, The Technical Cooperation

Program (TTCP), WPN/TP-1, KTA 1-29, August 1998

25. Ripple, G.R. and Mundie, T.G., Medical Evaluation of Nonfragment Injury

Effects in Armored Vehicle Live Fire Tests”, ADA233058, Walter Reed Army

Institute of Research (WRAIR), D.C., USA, 1989

lumbar load as relevant spinal injury metrics comparing ...comparing the use of dynamic response...

Documents